Non-Invasive Blood Glucose Prediction by Deep Learning System
Publish time: 2022-07-18 00:00:00
Non-Invasive Blood Glucose Prediction by Deep Learning System
Abstract
In this work, we use a non-invasive blood glucose detection method based on Photoplethysmography, and deep learning operations to estimate blood glucose value of patients. The accuracy of the blood glucose is achieved to 94.3% using our deep learning computing architecture with HbA1C value.
Technological Advantages
- High accuracy of Non-Invasive Blood Glucose Prediction
- Real time for Non-Invasive Blood Glucose Prediction
- Novel deep learning algorithm model performance

Figure1 The average training and testing loss of each model
Docket Number
26A-1101125
IP Status
TW I823501、Patent pending
Applications of Technology
- Non-Invasive Blood Glucose Prediction
- Personalized medicine
- Precision medicine
Inventor
Fu-Liang Yang; Justin Chu
Contact person
If you are interested in this technology, please fill in the information on the webpage under the manufacturer option (Request for Technology Licensing, Material Transfer, or Sponsored Research) on the website of this office

Non-Invasive Blood Glucose Prediction by Deep Learning System